A graphophonic investigation of beginning level texts.

摘要

This study attempted to provide a systematic framework for phonics instruction for beginning readers in literature-based classrooms based on relative frequency of phoneme-grapheme occurrences found in three distinct corpora. The first corpus contained an academic word list. The second corpus contained the running text from 363 books identified as first grade literature using the searchable online quiz database maintained by Renaissance Learning, Inc. (Renaissance Learning, 2009). The final corpus consisted of running text from 130 decodable readers that accompany Saxon Phonics 1: An Incremental Development (Simmons & Calvert, 2003). Each corpus was analyzed for graphophonic content in order to establish frequency distributions for 190 phoneme-grapheme correspondences. Instructional sequences were established for each corpus according to descending frequencies of the 190 correspondences. The instructional sequences were then statistically compared using a series of Spearman rank order correlations. It was found that a large significant correlation exists between the graphophonic distributions of the academic word list and the running text from first grade literature (rs = .80, p .05, N = 190), as well as between the running text of first grade literature and the running text from decodable phonics readers (rs = .955, p .05, N = 190). The conclusions supported by the findings are as follows: (a) an alternate sequence for teaching phoneme-grapheme correspondences is not supported based on frequency alone, (b) reading teachers adhering to an interactive approach to beginning reading instruction could theoretically use either literature or phonics text type to support early reading development, and (c) first graders need to be introduced to more phoneme-grapheme correspondences in order to be successful readers of first grade literature. The implications for practice which stemmed from these conclusions are twofold: (a) the leveling of texts should be fluid rather than stagnant, and (b) if reading development is dependent upon a student's ability to practice what has been taught and if the leveling of texts can only be done by human decision rather than by computer calculation, then teachers need expert training in the examination of curricular scope and sequences and matching texts to adopted curricula.